Sensitivity and specificity of anthropometrics for the prediction of diabetes in a biracial cohort

Abstract

Objective: To evaluate the ability of body mass index, waist circumference, waist-to-hip ratio, and combinations of these variables to discriminate individuals who will develop diabetes in adulthood.

Research methods and procedures: Data were from 45- to 64-year-old men and women who were members of the Atherosclerosis Risk in Communities cohort. The analysis sample consisted of 12,814 African American and white participants who were free of diabetes at baseline. Body mass index, waist circumference, waist-to-hip ratio, and diabetes incidence (defined as one glucose measure > or =126 mg/dL after fasting for at least 8 hours, one nonfasting glucose measure > or =200 mg/dL, and self-report of diabetes or report of taking medication for diabetes).

Results: 1515 new cases of diabetes were identified over the 9-year follow-up. Areas under receiver operating characteristic curves ranged from 0.66 to 0.73 for single measures. The curves were smooth, with no indication of a threshold. Waist tended to have the highest receiver operating characteristic statistic in all groups, but differences were small.

Discussion: The three anthropometric indices tested were approximately equivalent in their ability to predict diabetes. Sensitivity and specificities differed among ethnic and gender groups.